DNA methylation is an important epigenetic modification that plays crucial roles in multiple biological processes. Technical advance, especially a variety of deep sequencing-based techniques, have made it possible to monitor DNA methylome changes at a single-base resolution. However, how to interpret the epigenetic information encoded by the fast accumulating methylome data is still challenging. DNA methylation is traditionally considered to disrupt the interactions between transcription factors (TFs) and cis-regulatory regions and thus, silence the expression of downstream target genes. However, recent studies, especially our recent discoveries, suggest that many TFs and co-factors preferentially bind to methylated DNA motifs and, in some cases, transactivate downstream gene expression, challenging the current paradigm of DNA methylation in transcription regulation. Identification of comprehensive sets of functional methylation sites and their interacting partners will greatly expand the protein-DNA interaction landscape in a new direction and promise significant advances in the understanding of the biological roles of DNA methylation. To achieve these goals, we propose four specific aims in this R01 application. First, we will survey all possible 8-base DNA sequence combinations to identify methylated sequences that can be recognized by human TFs. A pool of methylated DNA motifs will be probed on the human TF protein microarrays and the DNA fragments that are captured by the proteins on the microarrays will be recovered and their sequences determined with deep-sequencing. Second, we will predict which of these methylated motifs are likely to play a role in gene regulation and interact with proteins. Those 8-mer sequences that are statistically enriched in the recovered population will be mapped to the available methylomes and examine whether they overlap with the known regulatory regions. The qualified motifs will be synthesized and individually probed on the protein microarrays to identify their binding partners. Third, we will predict the protein domains that are responsible fo methylated DNA binding. The sequences from the same TF subfamilies will be compared and the positions that can best separated the proteins with and without methylated binding activities will be the candidates for methylation binding. The prediction will be tested by site-directed mutagenesis coupled with gel shift and cell-based luciferase assays. Finally, we will use both in vitro and in vivo models of mammalian axon regeneration to investigate the physiological roles of newly identified mCpG-dependent TF-DNA interactions. The positive results provided by this Aim will not only reveal novel epigenetic mechanisms of mammalian axon regeneration, but also provide proof-of-concept evidence that mCpG-dependent TF-DNA interactions are physiological regulators of gene expression.